COVID-19, which is caused by 2019 novel coronavirus (SARS-CoV-2), was unexpected challenge to the Globe. After taking hundreds of thousands of lives, millions of people are infected by this virus worldwide. Since the first covid-19 case in Ethiopia was detected on 21-April-2020, more than 470,000 people have been infected by the pandemic with a total 7509 reported deaths until 21 April 2022. In this study some models based on ARIMA time series methodology for the forecasting of COVID-19 daily cases and deaths in Ethiopia was used. The data set covers from 13 March 2020 to 15 October 2020 extracted from the health ministry page.The study reveals that, ARIMA (3, 1, 2) is effective in modeling and forecasting daily Covid-19 cases. On the other hand ARIMA (4, 1, 3) is accurate in forecasting daily deaths from Covid-19. I have forecasted the COVID-19 data set for next 30 days. The forecasted values are in good agreement with real ones for both daily cases and deaths form COVID-19. ARIMA models can provide accurate COVID-19 predictions despite limited data points. Not only are they effective, they are also an easy way to predict COVID-19 trends based on open access data. Therefore, by adding new data points, we can be confident that the ARIMA model can be used to make accurate and reliable predictions of his COVID-19 cases for each day since October 2020.
